{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/adversarial-attack-on-graph-structured-data","title":"Adversarial Attack on Graph Structured Data","arxiv_id":"1806.02371","date":"2018-06-06","proceeding":"ICML 2018 7","authors":["Hanjun Dai","Hui Li","Tian Tian","Xin Huang","Lin Wang","Jun Zhu","Le Song"],"abstract":"Deep learning on graph structures has shown exciting results in various\napplications. However, few attentions have been paid to the robustness of such\nmodels, in contrast to numerous research work for image or text adversarial\nattack and defense. In this paper, we focus on the adversarial attacks that\nfool the model by modifying the combinatorial structure of data. We first\npropose a reinforcement learning based attack method that learns the\ngeneralizable attack policy, while only requiring prediction labels from the\ntarget classifier. Also, variants of genetic algorithms and gradient methods\nare presented in the scenario where prediction confidence or gradients are\navailable. We use both synthetic and real-world data to show that, a family of\nGraph Neural Network models are vulnerable to these attacks, in both\ngraph-level and node-level classification tasks. We also show such attacks can\nbe used to diagnose the learned classifiers.","url_abs":"http://arxiv.org/abs/1806.02371v1","url_pdf":"http://arxiv.org/pdf/1806.02371v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"adversarial-attack-on-graph-structured-data","repo_url":"https://github.com/Hanjun-Dai/graph_adversarial_attack","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"adversarial-attack","task_name":"Adversarial Attack"},{"task_slug":"graph-neural-network","task_name":"Graph Neural Network"},{"task_slug":"reinforcement-learning","task_name":"Reinforcement Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1806.02371","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}